Dietary Selenium Intake and Type-2 Diabetes: A Cross-Sectional Population-Based Study on CUME Project

Background and Aim: Previous studies have suggested that the specific association between selenium (Se) and diabetes remains unclear. This study aimed to investigate the association between dietary Se and type-2 diabetes (T2D) in the Brazilian cohort [Cohort of Universities of Minas Gerais (CUME)]. Methods and Results: This cross-sectional study was conducted with a large sample comprising 4,106 participants of the CUME project, a concurrent open cohort restricted to a highly educated population group, composed of graduates of federal institutions of higher education located in the State of Minas Gerais, Brazil. Data on socioeconomic and dietary characteristics, as well as anthropometric measures, were collected from each subject for analysis. The sample was classified into energy-adjusted tertiles of dietary Se intake (μg/day). Differences in the continuous data were evaluated by the Kruskal–Wallis H-test (abnormal data), and the χ(2)-test assessed differences in qualitative data. As there was no relationship between T2D and Se intake in the bivariate analysis, multivariate analysis was not performed. The prevalence of T2D in the studied population was 2.8%. The mean age was 36 years. Regarding gender, 1,209 are males and 2,807 are females. Among females, the mean Se intake was 165.12 μg/day and the mean intake was 157.4 μg/day. Among males, it was 168.4 μg/day. Significant differences were observed across all Se intake tertiles in terms of age, gender, activity level, alcohol intake, energy intake, sugar, carbohydrates, lipids, fiber, and energy-adjusted meat intake. However, no significant differences were observed across all Se intake tertiles in terms of BMI, smoking status, and T2D. The results indicated that there was no significant association between dietary Se intake and the prevalence of T2D. Conclusion: Dietary Se intake was not associated with the prevalence of T2D, despite the high intake of this micronutrient in the sample. These results contradict studies that identified the association between Se intake and T2D, with values of Se intake much lower than those observed in this study. Thus, this relationship seems to remain controversial.